Information Technology (PhD)
Permanent URI for this collection
Browse
Recent Submissions
Item 5W1H-Aware Approach for Retrieving Semantically Rich Multidimensional Events in Social Media Ecosystem(Addis Ababa University, 2024-08) Siraj Mohammed; Richard ChbeirNowadays, multimedia digital ecosystems (e.g., social media sites) become a great source of user-contributed multimedia documents for many types of real-world events. Social media documents about events are multimedia (e.g., image, video, text, and others), contain multiple features (describing different characteristics in it such as 5W1H), and come from multiple sources. Such events can be used to construct Event Knowledge Graph (EKG) which is basic to retrieve semantically rich multidimensional events. However, social media documents cannot be used directly to construct such a knowledge graph. First, social media documents must be represented in order to detect multimedia events as well as their different types of semantic relationships. To achieve these tasks, it is necessary to carry out preliminary event-related tasks, such as detecting, linking, and representing events. By doing these, we can provide an event search API that presents a concise summary of events focused on temporal, spatial, semantic, and participant aspects. For this, we proposed a novel 5W1H-aware framework consisting of six modules. Each of these modules uses 4W elements. More specifically, we first represent social media documents that have been used to detect real-world events with their 4W elements based on event-only descriptive features. The detected events are used to identify the three main event relations categories (such as, spatial, temporal, and sematic) and many relation types under these categories by comparing dimensions over multimedia events. We then used a graph database to store event detection outputs as nodes and relationship identification outputs as edges to construct an Event Knowledge Graph (EKG). Finally, we integrated the EKG with an event retrieval API to retrieve events. Each of these event-related tasks were evaluated using various datasets. For instance, the effectiveness of an incremental event detection algorithm was evaluated using the MediaEval 2013 dataset. The algorithm achieved an NMI score of 0.9914 and an F-score value of 0.9928 when the feature weights were assigned as 0.40 for participant, 0.30 for temporal, 0.35 for spatial, and 0.45 for semantic. Manually crafted spatial and temporal datasets are also used for evaluating the effectiveness of event relationship identification algorithm. Finally, the effectiveness of the developed EKG was evaluated through its downstream tasks, like retrieving similar nodes, whereas the event search API was evaluated via PageRank and search result relevance analysis. Results from the experiments showed that the proposed approach was more effective compared with alternative solutions.Item Reducing the Impact of Crowd Geo-localization on Battery Life by Proposing Novel Information Distribution Algorithm(Addis Ababa University, 2017-03) Samuel Asferaw; Jean-Marc (Professor); Georges Da-Costa (Professor); Dejene Ejigu (PhD)Geo-localization is an inevitable challenge and problem which has been studied for many years when dealing with wireless nodes as they are equipped with limited life batteries. Emergence of ‘crowd of wireless node users’ across the world (especially in big cities and emerging big cities in developing countries) and increasing sensor network applications are exacerbating the problem. Existing wireless node localization systems depend on power hungry systems like GPS. Advances in battery design have been slow and difficult; therefore, it is essential for alternative strategies to be employed to realize the goals of reducing the impact of crowd geo-localization on battery life. Based on the type of information required for localization, localization methods can be divided into two categories: range-based and range-free protocols. Since range-based methods demand battery hungry dedicated ranging components, range-free methods are explored. This thesis focuses on the investigation of novel smart algorithms for crowd geo-localization. We have proposed the following three but related smart novel rang-free localization algorithms for wireless networks: •Centre of the Smallest Communication Overlap Polygon (CSCOP) localization algorithm •Selective Anchor Nodes CSCOP localization algorithm •Immune to Radio Range Difference (IRRD) localization algorithm To evaluate the accuracy and reliability of the proposed algorithms, we have conducted extensive MATLAB simulations and comparisons with other state-of-the-art related work on: average, minimum, and maximum location error in a scenario of both sparse and dense crowd. Moreover, to see the probability of error distribution of the proposed algorithms, probability distribution of location error is developed using MATLAB normal fit function. We have also investigated computational complexity analysis of the algorithm using “Big O” notation along with detailed elementary operations used in the algorithms. Results obtained from our experiments show our algorithms significantly out-perform other related work in the domain in case of both sparse and dense crowd geo-localization. When we compare localization accuracy performance of our algorithms in case of sparse and dense crowd scenarios, they perform better in dense crowd scenario. For example, the average location error of our CSCOP algorithm in case of dense crowd with 30 anchor nodes involved is 3.7176 (in % radio range) which means 0.7435m error while its error in sparse crowd with 8 anchor nodes is 8.3552 (in % radio range) i.e., 1.6710m in our simulation scenario, where communication range is 20m. The algorithms, in addition to estimating reliable location, also define the smallest communication overlap polygon (SCOP) which can be applied in search and rescue operations as margins where the node situated. Moreover, unlike other related works which require at least three anchor nodes, the proposed algorithms work starting with two anchor nodes. Furthermore, unlike other related algorithms, our Immune to Radio Range Difference algorithm breaks the traditional assumption that says “anchor nodes have the same radio range”, by working in both homogeneous and heterogeneous radio ranges which is a break through finding in the domain area. Although, this work focuses on the crowd networks, applications of the proposed geo-localization algorithms could range to all wireless networks from cellular to sensor.Item Hybrid Model for Amharic Sentiment Classification(Addis Ababa University, 2022-01) Girma Neshir; Andreas Rauber (Professor)Amharic is a less-resourced language, which lacks a standard dictionary, stemmer, language detector, subjectivity detection, negation handling, Amharic sentiment lexicon and annotated Amharic corpora to carry out sentiment classification in social media texts. This research focuses on sentiment analysis of Amharic texts. The first part of the research is to generate most of these required resources and the second part of the research deals with enhancing performance of sentiment classification using proposed approaches. In this research, four categories of corpora are prepared: general corpora (category I), annotated corpora (category II), lexical resources (category III) and pre-trained models (category IV). The annotated corpora, such as Facebook comments of GCAO (2,871), PMO (6,637), EBC (2,444) and Zemen YouTube Comments(1,440) are used for building and evaluating Amharic sentiment classification approaches. To remedy the problems of sentiment analysis of an under-resourced language (Amharic in this case), Amharic sentiment lexicons are generated using dictionary based and corpus based approaches. Using a dictionary based approach, SO-CAL (5,681) and SWN (13,677) are generated from English sentiment lexicons (using category III). Using corpus based approach, Amharic sentiment seeds are expanded to generate Amharic sentiment lexicon from Amharic corpora (using category I). At the threshold of 500, the generated lexicon has a size of 8,132. The generated lexicons are evaluated in terms of subjectivity detection, coverage, agreement and accuracy by comparison with the manual lexicon (baseline). The generated lexicons are used for subjectivity detection and negation handling. For sentiment classification (SC) of text on a topic, supervised, ensemble methods and BERT transfer learning are proposed, built, tuned and evaluated under small labeled observations (using category II). Finally, for enhancing the performance of Amharic sentiment classification, a hybrid model (i.e. voting, averaging and blender) is developed that combines the top performing classifiers of the earlier approaches. Experiments on the proposed hybrid models were done using category II annotated sentiment data sets. The results show that the proposed hybrid model (blender) has achieved performance gain as compared to SVM model (baseline) using the data sets. The complete sentiment classification system is showed in real-time and offline applications for detecting language, topics, subjectivity and prediction of sentiments of comments.Item Framework for Software Requirements Engineering Ethics(Addis Ababa University, 2025-05) Seblewongel Esseynew; Nuno M. Garcia (Professor)Ethics has an extensive array of uses throughout many fields of study and in diverse circumstances. Ethics significantly impacts decisions made in requirements engineering. Requirements engineering is a fundamental process in software development projects. At the same time, it is a difficult phase and exposes to many ethical problems. In software engineering, many well-known codes of ethics address a wide range of issues that are significant for software engineers with different perspectives. However, it is found to be challenging to follow and implement them in the day-to-day work of software engineers. This research aimed to study ethical issues raised in requirements engineering processes. In this research, we investigated requirements engineering ethical concerns. Data were collected from seven case software companies using interviews and focus group discussions. Themantic qualitative analysis is employed to analyse the collected data. The analysis result revealed that companies lack industry practices, use of professional code of conduct standards, methods, or any other ethical guidelines to be followed during requirements engineering. Based on the analysis, one hundred seventy six codes, thirty four categories and five themes emerged. The five main themes include importance, practice, fundamental ethical concerns, management-related problems, and methods. In this research, we proposed a framework with five basic components for software requirements engineering ethics to address critical ethical concerns which were screened and identified during the analysis per their imprance to achieve the research questions. These components are Ethics for Knowledge Gap, Ethics for Requirements Identification, Ethics for Requirements Quality, Ethics for Unwillingness, and Ethics for doing Unpermitted Activities. Each of them gives specific pieces of advice that can be solutions to address the identified requirements engineering ethical concerns. The validity and reliability of the proposed framework were evaluated using an expert evaluation involving five-point Likert scale questionnaires and focus group discussions. Additionally, open-ended questionnaires are used. The experts are purposively selected from the software industries. The evaluation results indicate that the suggested framework is beneficial to use it as a daily guideline. It can be used as an advising guide for software engineers to minimize the occurrence of those identified ethical concerns during requirements engineering.Item Modeling the Knowledge Factors that Explain Onshore Information Systems Development Outsourcing Success in Ethiopia: a Client-Vendor Dyad(Addis Ababa University, 2021-07) Solomon Abebe; Alemayehu Molla (Professor); Temtim Assefa (PhD)Despite the increased organizational spending on Information Systems Development Outsourcing (ISDO), ensuring ISDO success and delivering business values from such initiative remains a challenge for business managers in a developing economy. In recent studies, there has been much interest in addressing the challenge from the antecedents and consequences of knowledge exchange between clients and vendors. However, a review of the literature on knowledge transfer and ISDO shows that most of the studies are (a) dominated by a one directional knowledge transfer from or to one of the outsourcing parties; (b) focused on examining ISDO success from either the clients' or vendors' perspective, but not both; (c) focused on offshore ISDO context of developed economies. Although these earlier studies are useful, the direct transferability of the findings to the context of developing economy countries such as Ethiopia is questionable. This is because ISDO arrangements in a developing economy are characterized by an immature ISDO market. Client organizations lack experience in how to govern outsourced IS/IT initiatives and outsourcing partners usually demonstrate a lower level of awareness about the value of knowledge due to unsupportive knowledge transfer culture. Therefore, research that investigates mutual ISDO success by addressing the limitations of prior research and capturing the significant context of developing economy through the lens of sound knowledge-related theories is needed. Thus, this study is set out to address the following research questions: (a) what are the key knowledge-related factors influencing onshore ISDO success? (b) how are these factors interrelated? and (c) to what extent do these factors contribute to onshore ISDO success? A review of the literature on ISDO, knowledge exchange, and ISDO success was conducted. Based on the findings from the literature, an exploratory study, and drawing from knowledge based view and social exchange theories, a knowledge-based onshore ISDO success conceptual framework was developed from the client-vendor perspective. This study was informed and guided by the positivist research paradigm. A mixed-methods research strategy particularly a sequential transformative mixed method was employed. The findings from the qualitative and exploratory study in Phase 1 were used as inputs in the development of the quantitative study in the second phase. Semi-structured interviews (10 project managers) and survey techniques were used to collect data in phases 1 and 2 respectively from the onshore ISDO practices in Ethiopia. SmartPLS 2.0 M3 statistical tool was used to analyze the survey data, assess the reliability and validity of the measurement model, and test the hypothesized relationships in the proposed conceptual model. The findings of the research show that onshore ISDO partners in a developing economy can deliver projects on time, increase their satisfaction with one another, and with the shared benefit and nurture long term partnership by (a) enhancing the utilization of knowledge transferred between vendors and clients; (b) developing their absorptive capacity, and (c) improving the effectiveness of knowledge exchange. To do so, the development of mutual disseminative capacity, mutual absorptive capacity, mutual learning intent, and mutual competence trust between clients and vendors play a significant role and in turn, indirectly contribute to ISDO success. Furthermore, the results of the study indicate that knowledge transfer and absorptive capacity contribute to knowledge utilization in onshore ISDO relationships. The findings also show that knowledge utilization partially mediates the relationships between knowledge transfer and ISDO success and mutual absorptive capacity and ISDO success. The study's results further suggest that mutual disseminative capacity, mutual absorptive capacity, and mutual learning intent facilitate knowledge transfer. Lastly, the results also confirm that in a bi-directional flow of knowledge, knowledge recipients’ absorptive capacity necessitates mutual disseminative capacity, mutual learning intent, and mutual competence trust of outsourcing partners. The research makes several contributions to ISDO research and practice. The overall findings indicate that taking the client-vendor dyads, instead of either a client or vendor perspective, offers a more complete picture of the knowledge dynamics that influence ISDO success. In terms of theory, the study contributes an onshore ISDO success framework to the ISDO literature in a developing economy. The utility of this framework is demonstrated as the research model explains 53.5% of the variance in onshore ISDO success. In addition, the study contributes sufficiently contextualized and research-ready instruments to operationalize constructs such as mutual knowledge dissemination, mutual absorptive capacity, mutual learning intent, mutual competence trust, knowledge transfer, and utilization. Together, the framework and instrument may encourage future research.Item Dynamic Ambidexterity for Balancing Standardization and Evolve-Ability in Information System Implementation in the Resource-Constrained Setting: the Case of DHIS2 Implementation in Public Health Care in Ethiopia(Addis Ababa University, 2023-06-26) Birkinesh W/Yohannes; Mark Gaynor (Professor)Health Information System (HIS) implementation involves inherently contradictory issues, standardization, and evolve-ability, that require different sets of information technology solutions and governance mechanisms (i.e. centralized/decentralized structure, controlled/autonomous process, top-down/bottom-up communication) which is difficult to materialize in practice. Contemporary studies indicate that organizations are shifting away from either/or strategies toward balancing that favors simultaneously achieving contradictory activities. However, theoretical and empirical research guiding how to organize and implement a balancing of standardization and evolve-ability is lacking. Furthermore, research has paid little attention to resource-constrained settings where sources of contradiction are immense due to a lack of technical and financial resources needed for HIS implementation. To address this shortcoming, this study seeks to study HIS implementation in the public healthcare sector of Ethiopia aiming to contribute to the development of an IT governance framework for understanding and addressing contradictory issues in the resource-constraining setting. Following an interpretive approach, the implementation of DHIS2, an open-source software developed at the University of Oslo, in the Ethiopian public health care setting was used as a case to explore the underlying logic of IT governance and IS implementation and its impacts. Managers, health, and IT professionals who were involved in the implementation were chosen as informants for this study to provide a deeper understanding of the IT governance mechanisms and their impact on the implementation. Institutional logic served as the primary analytical tool for this study and was used to identify common themes for stakeholders. Dynamic ambidexterity and boundary resource concepts became parts of analytical tools to explore the process of IS implementation at organizational and project levels as the study progressed from data collection, analysis, and the development of theoretical inferences. This study proposed an integrated ambidextrous implementation governance framework, drawn from institutional logics; dynamic ambidexterity, and boundary resource model; to complement the resource-chasing collaboration with major logics-collaboration, uncontrolled and controlled processes. This framework can be considered a major theoretical contribution that requires the central actor’s institutional distance to accommodate major collaboration with diverse underlying logic. This required enabling common visions, strategies, and mechanisms for sensing, seizing, and reconfiguring the emergent heterogeneous system implementation initiatives that lead to continuously and simultaneously achieving standardization and evolve-ability. This study is limited by data from one region and a specific system which might not adequately describe the complexity of a national HIS implementation. Furthermore, the researcher's role as an external participant would miss important events such as internal meetings where technical decision and discussion was made which should have been ideal to get rich information. Furthermore, the research indicates the importance of technological features to shape governance mechanisms. Hence further research can investigate the interrelationship between IT governance and the technology feature. Further research is needed to extend the scope of this study to multiple regions, and systems for a longer time to highlight new insights and to highlight the differences across regions and systems. These are limitations and potential areas for further research.Item Dynamic Ambidexterity for Balancing Standardization and Evolve-Ability in Information System Implementation in the Resource-Constrained Setting: the Case Of DHIA2 Implementation in Public Health Care in Ethiopia(Addis Ababa University, 2023-06-26) Birkinesh W/Yohannes; Mark Gaynor (Professor); Temtim Assefa (PhD)Health Information System (HIS) implementation involves inherently contradictory issues, standardization, and evolve-ability, that require different sets of information technology solutions and governance mechanisms (i.e. centralized/decentralized structure, controlled/autonomous process, top-down/bottom-up communication) which is difficult to materialize in practice. Contemporary studies indicate that organizations are shifting away from either/or strategies toward balancing that favors simultaneously achieving contradictory activities. However, theoretical and empirical research guiding how to organize and implement a balancing of standardization and evolve-ability is lacking. Furthermore, research has paid little attention to resource-constrained settings where sources of contradiction are immense due to a lack of technical and financial resources needed for HIS implementation. To address this shortcoming, this study seeks to study HIS implementation in the public healthcare sector of Ethiopia aiming to contribute to the development of an IT governance framework for understanding and addressing contradictory issues in the resource-constraining setting. Following an interpretive approach, the implementation of DHIS2, an open-source software developed at the University of Oslo, in the Ethiopian public health care setting was used as a case to explore the underlying logic of IT governance and IS implementation and its impacts. Managers, health, and IT professionals who were involved in the implementation were chosen as informants for this study to provide a deeper understanding of the IT governance mechanisms and their impact on the implementation. Institutional logic served as the primary analytical tool for this study and was used to identify common themes for stakeholders. Dynamic ambidexterity and boundary resource concepts became parts of analytical tools to explore the process of IS implementation at organizational and project levels as the study progressed from data collection, analysis, and the development of theoretical inferences. This study proposed an integrated ambidextrous implementation governance framework, drawn from institutional logics; dynamic ambidexterity, and boundary resource model; to complement the resource-chasing collaboration with major logics-collaboration, uncontrolled and controlled processes. This framework can be considered a major theoretical contribution that requires the central actor’s institutional distance to accommodate major collaboration with diverse underlying logic. This required enabling common visions, strategies, and mechanisms for sensing, seizing, and reconfiguring the emergent heterogeneous system implementation initiatives that lead to continuously and simultaneously achieving standardization and evolve-ability. This study is limited by data from one region and a specific system which might not adequately describe the complexity of a national HIS implementation. Furthermore, the researcher's role as an external participant would miss important events such as internal meetings where technical decision and discussion was made which should have been ideal to get rich information. Furthermore, the research indicates the importance of technological features to shape governance mechanisms. Hence further research can investigate the interrelationship between IT governance and the technology feature. Further research is needed to extend the scope of this study to multiple regions, and systems for a longer time to highlight new insights and to highlight the differences across regions and systems. These are limitations and potential areas for further research.